ECE444-PRA5: Deploying an ML APP to the Cloud

Latency Results

Here were the averages for each case in ms: Test Case Case 1 60.439620 Case 2 58.780115 Case 3 57.438099 Case 4 57.022431

Below is a boxplot visualizing the response times for the 100 requests:

API Endpoint

• Deployed URL: http://serve-sentiment-env.eba-gt2b3ixf.us-east-2.elasticbeanstalk.com/predict
• Method: POST
• Input: A JSON payload with a text field containing the news headline.

Example:

{
  "text": "Scientists confirm that the Earth is actually flat and not round."
}